INVESTIGATION Selective Sweeps Under Dominance and Inbreeding Matthew Hartfield*,†,‡,1 and Thomas Bataillon† *Department of Ecology and Evolutionary Biology, University of Toronto, Ontario M5S 3B2, Canada, †Bioinformatics Research Centre, Aarhus University, Aarhus 8000, Denmark, and ‡Institute of Evolutionary Biology, The University of Edinburgh, Edinburgh EH9 3FL, United Kingdom ORCID IDs: 0000-0001-6941-0647 (M.H.); 0000-0002-4730-2538 (T.B.) ABSTRACT A major research goal in evolutionary genetics is to uncover loci experiencing positive KEYWORDS selection. One approach involves finding ‘selective sweeps’ patterns, which can either be ‘hard sweeps’ Adaptation formed by de novo mutation, or ‘soft sweeps’ arising from recurrent mutation or existing standing variation. Dominance Existing theory generally assumes outcrossing populations, and it is unclear how dominance affects soft Self-fertilisation sweeps. We consider how arbitrary dominance and inbreeding via self-fertilization affect hard and soft Selective Sweeps sweep signatures. With increased self-fertilization, they are maintained over longer map distances due to Population reduced effective recombination and faster beneficial allele fixation times. Dominance can affect sweep Genetics patterns in outcrossers if the derived variant originates from either a single novel allele, or from recurrent mutation. These models highlight the challenges in distinguishing hard and soft sweeps, and propose methods to differentiate between scenarios. Inferring adaptive mutations from nucleotide polymorphism data are a Classic hitchhiking models consider ‘hard’ sweeps, where the com- major research goal in evolutionary genetics, and has been subject to mon ancestor of an adaptive allele occurs after the onset of selection extensive modeling work to determine the footprints they leave in (Hermisson and Pennings 2017). Recent years have seen a focus on genome data (Stephan 2019). The earliest models focused on a scenario ‘soft’ sweeps, where the most recent common ancestor of a beneficial where a beneficial mutation arose as a single copy before rapidly fixing. allele appeared before it became selected for (reviewed by Barrett and Linked neutral mutations then ‘hitchhike’ to fixation with the adaptive Schluter (2008); Messer and Petrov (2013); Hermisson and Pennings variant, reducing diversity around the selected locus (Maynard Smith and (2017)). Soft sweeps can originate from beneficial mutations being Haigh 1974; Kaplan et al. 1989). Hitchhiking also increases linkage dis- introduced by recurrent mutation at the target locus (Pennings and equilibrium in regions flanking the selected site, by raising the haplotype Hermisson 2006a, b), or originating from existing standing variation carrying the selected allele to high frequency (Thomson 1977; Innan and that was either neutral or deleterious (Orr and Betancourt 2001; Nordborg 2003; McVean 2007). These theoretical expectations have Innan and Kim 2004; Przeworski et al. 2005; Hermisson and Pen- spurred the creation of summary statistics for detecting sweeps, usually nings 2005; Wilson et al. 2014; Berg and Coop 2015; Wilson et al. based on finding genetic regions exhibiting extended haplotype homo- 2017). A key property of soft sweeps is that the beneficial variant is zygosity (Sabeti et al. 2002; Kim and Nielsen 2004; Voight et al. 2006; present on multiple genetic backgrounds as it sweeps to fixation, so Ferrer-Admetlla et al. 2014; Vatsiou et al. 2016), or an increase in high different haplotypes may carry the derived allele. This property is often frequency derived variants (Fay and Wu 2000; Kim and Stephan 2002; used to detect soft sweeps in genetic data (Peter et al. 2012; Vitti et al. 2013; Nielsen 2005; Boitard et al. 2009; Yang et al. 2018; Fujito et al. 2018). Garud et al. 2015; Garud and Petrov 2016; Schrider and Kern 2016; Sheehan and Song 2016; Harris et al. 2018a; Kern and Schrider 2018; Harris and DeGiorgio 2018, 2019). Soft sweeps have been reported in Copyright © 2020 Hartfield, Bataillon Drosophila (Karasov et al. 2010; Garud et al. 2015; Garud and Petrov doi: https://doi.org/10.1534/g3.119.400919 2016; Vy et al. 2017), humans (Peter et al. 2012; Schrider and Kern 2017; Manuscript received November 18, 2019; accepted for publication January 18, 2020; published Early Online January 23, 2020. Laval et al. 2019), maize (Fustier et al. 2017), Anopheles mosquitoes (Xue This is an open-access article distributed under the terms of the Creative et al. 2019), and pathogens including Plasmodium falciparum (Anderson Commons Attribution 4.0 International License (http://creativecommons.org/ et al. 2016) and HIV (Pennings et al. 2014; Williams and Pennings 2019). licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction Yet determining how extensive soft sweeps are in nature remains a in any medium, provided the original work is properly cited. Supplemental material available at figshare: https://doi.org/10.25387/ contentious issue (Jensen 2014; Harris et al. 2018b). g3.11687949. Up to now, there have only been a few investigations into how 1Corresponding author: E-mail: m.hartfi[email protected] dominance affects sweep signatures. In a simulation study, Teshima and Volume 10 | March 2020 | 1063 Przeworski (2006) explored how recessive mutations spend long diversity present following a sweep from either a de novo mutation, or from periods of time at low frequencies, increasing the amount of re- standing variation. We also determine the number of segregating sites and combination that acts on derived haplotypes, weakening signatures the site frequency spectrum, while comparing results to an alternative soft- of hard sweeps. Fully recessive mutations may need a long time to sweep model where adaptive alleles arise via recurrent mutation. Note that reach a significantly high frequency to be detectable by genome we focus here on single sweep events, rather than characterizing how scans (Teshima et al. 2006). Ewing et al. (2011) have carried out sweeps affect genome-wide diversity (Elyashiv et al. 2016; Campos a general mathematical analysis of how dominance affects hard et al. 2017; Booker and Keightley 2018; Rettelbach et al. 2019). sweeps, finding that recessive beneficial mutations have markedly different signatures compared to those with other dominance values. METHODS Yet the impact of dominance on soft sweeps has yet to be explored in depth. Model outline In addition, existing models have so far focused on randomly We consider a diploid population of size N (carrying 2N haplotypes in mating populations, with haplotypesfreelymixingbetweenindi- total). Individuals reproduce by self-fertilization with probability s, viduals over generations. Different reproductive modes alter how and outcross with probability 1 2 s. A derived allele arises at a locus, alleles are inherited, potentially changing the hitchhiking effect. and we are interested in determining the population history of neutral Self-fertilization, where male and female gametes produced from the regions that are linked to it, with a recombination rate r between same individual can fertilize one another, can alter adaptation rates them. We principally look at the case where the beneficial allele arises and selection signatures (Hartfield et al. 2017). This mating system from previously–neutral standing variation, and subsequently look at is prevalent among angiosperms (Igic and Kohn 2006), some ani- a sweep arising from recurrent mutation. The derived allele initially mals (Jarne and Auld 2006) and fungi (Billiard et al. 2011). As the segregates neutrally for a period of time, then becomes advantageous effects of dominance and self-fertilization become strongly inter- with selective advantage 1 þ hs when heterozygous and 1 þ s when twined, it is important to consider both together. Dominant mu- homozygous, with 0 , h , 1 and s . 0. We further assume that the tations are more likely to fix than recessive ones in outcrossers, as population size is large and selection is large enough so that the they have a higher initial selection advantage (Haldane 1927). Yet beneficial allele’s change in frequency can be modeled deterministi- recessive alleles can fix more easily in selfers than in outcrossers as cally (i.e., Nehs 1 and 1=Ne s 1). Table 1 lists the notation homozygote mutations are created more rapidly (Charlesworth used in the analysis. 1992; Glémin 2012). Furthermore, a decrease in effective recom- Our goal is to determine how the spread of the derived, adaptive bination rates in selfers (Nordborg et al. 1996; Nordborg 2000; allele affects genealogies at linked neutral regions. For a sweep Charlesworth and Charlesworth 2010) can interfere with selec- originating from standing variation, we follow the approach of tion acting at linked sites, making it likelier that deleterious mu- Berg and Coop (2015) and, looking backward in time, break down tations hitchhike to fixation with adaptive alleles (Hartfield and the selected allele history into two phases. In the recent past is the Glémin 2014), or that rare mutations are lost by drift due to ‘sweep phase’ where the derived allele was selectively favored, with competition between adaptive mutations (Hartfield and Glémin its frequency decreasing from 1 to p0. Prior to that phase is the 2016). ‘standing phase’, which assumes that the derived allele is present at In a constant-sized population, beneficial mutations can be less likely an approximate fixed frequency p0. During both phases, a pair of to fix from standing variation (either